Book contents
- Frontmatter
- Contents
- List of contributors
- Preface
- Introduction
- Inference and learning in latent Markov models
- Part I State space methods for neural data
- State space methods for MEG source reconstruction
- Autoregressive modeling of fMRI time series: state space approaches and the general linear model
- State space models and their spectral decomposition in dynamic causal modeling
- Estimating state and parameters in state space models of spike trains
- Bayesian inference for latent stepping and ramping models of spike train data
- Probabilistic approaches to uncover rat hippocampal population codes
- Neural decoding in motor cortex using state space models with hidden states
- State space modeling for analysis of behavior in learning experiments
- Part II State space methods for clinical data
- index
- References
Autoregressive modeling of fMRI time series: state space approaches and the general linear model
from Part I - State space methods for neural data
Published online by Cambridge University Press: 05 October 2015
Edited by
Book contents
- Frontmatter
- Contents
- List of contributors
- Preface
- Introduction
- Inference and learning in latent Markov models
- Part I State space methods for neural data
- State space methods for MEG source reconstruction
- Autoregressive modeling of fMRI time series: state space approaches and the general linear model
- State space models and their spectral decomposition in dynamic causal modeling
- Estimating state and parameters in state space models of spike trains
- Bayesian inference for latent stepping and ramping models of spike train data
- Probabilistic approaches to uncover rat hippocampal population codes
- Neural decoding in motor cortex using state space models with hidden states
- State space modeling for analysis of behavior in learning experiments
- Part II State space methods for clinical data
- index
- References
Summary
Introduction
In this chapter we discuss predictive modeling of series time obtained by functional magnetic resonance imaging (fMRI), representing an important case of spatiotemporal data. Following its development in the early 1990s, fMRI has become a well established approach to investigating brain activity in vivo (Huettel et al. 2004), providing temporally and spatially resolved recordings of the “blood oxygen level dependent” (BOLD) signal of neural tissue. fMRI time series consist of a temporal sequence of scans of the brain and the surrounding space (discretized into voxels), such that the resulting data sets may be stored as vector time series.
The practical work with fMRI time series poses considerable challenges in many aspects, including the huge dimensionality of the data, which usually is recorded from several 10 of voxels, and the plethora of artifacts and contaminations disturbing the data (Strother 2006). Further difficulties arise from the low temporal sampling frequency, typically well below 1 Hz, and the indirect relationship between the BOLD signal and the underlying neural processes.
Currently available approaches to fMRI time series analysis may be broadly classified into three groups:
• exploratory methods, such as cluster analysis (Goutte et al. 1999), principal component analysis (PCA) (Anderson et al. 1999) and independent component analysis (ICA) (McKeown 2000);
• massively univariate (voxel-wise) regression methods, implemented in software packages such as statistical parametric mapping (SPM) (Friston et al. 1994), the FMRIB software library (FSL) (Smith et al. 2004), or the analysis of functional Neuroimages (AFNI) package (Cox 1996);
• generative dynamic models, based on specific assumptions regarding the properties of the underlying neural masses and the biophysical processes which produce the experimental data; as examples we mention dynamic causal modeling (DCM) (Friston et al. 2003) and the hemodynamic state space model (SSM) of Riera et al. (2004a).
Among these three groups of methods, the third may be interpreted as an example of predictive modeling, while for the second group this is possible only in a very limited sense, and essentially impossible for the first group.
- Type
- Chapter
- Information
- Advanced State Space Methods for Neural and Clinical Data , pp. 79 - 113Publisher: Cambridge University PressPrint publication year: 2015
References
& , eds (1999). The Practice of Time Series Analysis, Berlin: Springer.
& (1988). Statistical Analysis and Control of Dynamic Systems, Dordrecht: Kluwer.
, & (1999). Principal component analysis of the dynamic response measured by fMRI: a generalized linear systems framework. Journal of Magnetic Resonance Imaging 17, 795–815.CrossRefGoogle Scholar
(1987). State Space Modeling of Time Series,Berlin: Springer.
, , , , & (2014). A subspace pursuit-based iterative greedy hierarchical solution to the neuromagnetic inverse problem. Neuroimage 87, 427–443.CrossRefGoogle Scholar
(2006). Pattern Recognition and Machine Learning, New York: Springer.
& (1998). Dynamic changes in effective connectivity characterised by variable parameter regression and Kalman filtering. Human Brain Mapping 6, 403–408.3.0.CO;2-9>CrossRefGoogle Scholar
, , , , , , , & (1996). Statistical methods of estimation and inference for functional MR image analysis. Magnetic Resonance in Medicine 35, 261–277.CrossRefGoogle Scholar
& (1999). Kalman Filtering: with Real-time Applications, 3rd edn, Berlin: Springer.
(1996). AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical Research 29, 162–173.CrossRefGoogle Scholar
, , , , & (2006). Dynamical causal modeling of evoked responses in EEG and MEG. Neuroimage 30, 1255–1272.CrossRefGoogle Scholar
& (2001). Time Series Analysis by State Space Methods, New York: Oxford University Press.
(1987). Practical Methods of Optimization, 2nd edn, New York: Wiley.
, , , , & (1998b). Eventrelated fMRI: characterising differential responses. Neuroimage 7, 30–40.Google Scholar
, , , & (1995). Characterizing evoked hemodynamics with fMRI. Neuroimage 2, 157–165.CrossRefGoogle Scholar
, & (1994). The analysis of functionalMRI time series. Human Brain Mapping 1, 153–171.CrossRefGoogle Scholar
, , & (1998a). Nonlinear event-related responses in fMRI. Magnetic Resonance in Medicine 39, 41–52.Google Scholar
, , & (2006). Whitening as a tool for estimating mutual information in spatiotemporal data sets. Journal of Statistical Physics 124, 1275–1315.CrossRefGoogle Scholar
, , , , , & (2010). Optimal HRF and smoothing parameters for FMRI time series within an autoregressive modeling framework. Journal of Integrative Neuroscience 9, 439–452.CrossRefGoogle Scholar
, , , & (2004). A solution to the dynamical inverse problem of EEG generation using spatiotemporal Kalman filtering. Neuroimage 23, 435–453.CrossRefGoogle Scholar
(1999). Deconvolution of impulse response in event-related BOLD fMRI. Neuroimage 9, 416–429.CrossRefGoogle Scholar
, & (2001). Bayesian modeling of the hemodynamic response function in BOLD fMRI. Neuroimage 14, 140–148.CrossRefGoogle Scholar
& (2001). Kalman Filtering: Theory and Practice Using MATLAB, New York: Wiley-Interscience.
, & (2003). Multivariate autoregressive modelling of fMRI time series. Neuroimage 19, 1477–1491.CrossRefGoogle Scholar
(2000). Parametric modelling of functional magnetic resonance imaging data. Ph.D. thesis, Department of Theoretical Statistics, University of Aarhus, Aarhus.
& (2006). Decoding mental states from brain activity in humans. Nature Reviews Neuroscience 7, 523–534.CrossRefGoogle Scholar
, & (2004). Functional Magnetic Resonance Imaging, Sunderland: Sinauer Associates.
(1968). An innovations approach to least-squares estimation – part I: Linear filtering in additive white noise. IEEE Transactions on Automatic Control 13, 646–655.CrossRefGoogle Scholar
(1980). Linear Systems, Englewood Cliffs, NJ: Prentice-Hall.
(1960). A new approach to linear filtering and prediction problems. Journal of Basic Engineering 82, 35–45.CrossRefGoogle Scholar
& (1997). Non-linear Fourier time series analysis for human brain mapping by functional magnetic resonance imaging. Applied Statistics 46, 1–29.CrossRefGoogle Scholar
, , & (1997). Time series analysis in the time domain and resampling methods for studies of functional magnetic resonance brain imaging. Human Brain Mapping 5, 168–193.3.0.CO;2-1>CrossRefGoogle Scholar
, , , & (2014). Variational Bayesian causal connectivity analysis for fMRI. Frontiers in Neuroinformatics 8.CrossRefGoogle Scholar
(1993). Introduction to Multiple Time Series Analysis, 2nd edn, Berlin: Springer.
(1982). Vision: a Computational Investigation into the Human Representation and Processing of Visual Information, San Francisco, CA: W.H. Freeman.
(2000). Detection of consistently task-related activation in fMRI data with hybrid independent component analysis. Neuroimage 11, 24–35.CrossRefGoogle Scholar
& (1965). A simplex method for function minimization. Computer Journal 7, 308–313.CrossRefGoogle Scholar
& (1972). Generalized linear models. Journal of the Royal Statistical Society 135, 370–384.CrossRefGoogle Scholar
(1992). A bridge between nonlinear time series models and nonlinear stochastic dynamical systems: a local linearization approach. Statistica Sinica 2, 113–135.Google Scholar
(2012). Time Series Modeling of Neuroscience Data, London: Chapman & Hall/CRC Press.
& (2001). An innovation approach to non-gaussian time series analysis. Journal of Applied Probability 38, 78–92.CrossRefGoogle Scholar
, & (2005a). Bilinear dynamical systems. Philosophical Transactions of the Royal Society, Series B 360, 983–993.Google Scholar
, & (2003). Variational bayesian inference for fMRI time series. Neuroimage 19, 727–741.CrossRefGoogle Scholar
, & (2005b). Bayesian fMRI time series analysis with spatial priors. Neuroimage 24, 350–362.Google Scholar
(1990). Stochastic Integration and Differential Equations, Berlin: Springer-Verlag.
, , & (2001). Locally regularized spatiotemporal modeling and model comparison for functional MRI. Neuroimage 14, 912–923.CrossRefGoogle Scholar
& (1995). Linear Models: Least Squares and Alternatives, Berlin: Springer.
, & (1965). Maximum likelihood estimates of linear dynamic systems. American Institute of Aeronautics and Astronautics (AIAA) Journal 3, 1445–1450.CrossRefGoogle Scholar
, , , , , & (2004b). fMRI activation maps based on the NN-ARX model. Neuroimage 23, 680–697.Google Scholar
, , , , , & (2004a). A state-space model of the hemodynamic approach: non-linear filtering of BOLD signal. Neuroimage 21, 547–567.Google Scholar
& (1982). An approach to time series smoothing and forecasting using the em algorithm. Journal of Time Series Analysis 3, 253–264.CrossRefGoogle Scholar
, , , , , , , , , , , , , , , & (2004). Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 23, 208–219.CrossRefGoogle Scholar
, , & (2001). A signal estimation approach to functional MRI. IEEE Transactions on Medical Imaging 20, 26–35.CrossRefGoogle Scholar
(2006). Evaluating fMRI preprocessing pipelines. IEEE Engineering in Medicine and Biology Magazine 25, 27–41.CrossRefGoogle Scholar
(1995). Tank model. In , ed., Computer Models of Watershead Hydrology, Highlands Ranch, CO: Water Resources Publications, pp. 165–214.
, , , , , & (2005). Estimating brain functional connectivity with sparse multivariate autoregression. Philosophical Transactions of the Royal Society, Series B 360, 969–981.CrossRefGoogle Scholar
, , & (2010). Efficient Bayesian multivariate fMRI analysis using a sparsifying spatio-temporal prior. Neuroimage 50, 150–161.CrossRefGoogle Scholar
(1949). The Extrapolation, Interpolation and Smoothing of Stationary Time Series with Engineering Applications, New York: Wiley.
, & (2004). Constrained linear basis sets for HRF modelling using variational Bayes. Neuroimage 21, 1748–1761.CrossRefGoogle Scholar
& (1995). Analysis of fMRI time series revisited – again. Neuroimage 2, 173–181.CrossRefGoogle Scholar
, , , , , & (2002). A general statistical analysis for fMRI data. Neuroimage 15, 1–15.CrossRefGoogle Scholar